mlsanityData to 14 Oct 2026

Provable Fairness for Neural Network Models using Formal Verification

arXiv 2212.08578 ↗First posted 16 Dec 2022cs.LG cs.CYcited by 6

Not on the map. Why?

Abstract (arXiv)

Machine learning models are increasingly deployed for critical decision-making tasks, making it important to verify that they do not contain gender or racial biases picked up from training data. Typical approaches to achieve fairness revolve around efforts to clean or curate training data, with post-hoc statistical evaluation of the fairness of the model on evaluation data. In contrast, we propose techniques to prove fairness using recently developed formal methods that verify properties of neural network models.Beyond the strength of guarantee implied by a formal proof, our methods have the advantage that we do not need explicit training or evaluation data (which is often proprietary) in order to analyze a given trained model. In experiments on two familiar datasets in the fairness literature (COMPAS and ADULTS), we show that through proper training, we can reduce unfairness by an average of 65.4% at a cost of less than 1% in AUC score.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 18 of its 29 references that are in the map. First posted on arXiv 16 Dec 2022.

Builds on

18 of 29 references.

References · 18

  1. 2019A Survey on Bias and Fairness in Machine Learningcited by 1,917
  2. 2017Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networkscited by 744
  3. 2014Certifying and removing disparate impactcited by 715
  4. 2020Fairness in Machine Learning: A Surveycited by 370
  5. 2018AI Fairness 360: An Extensible Toolkit for Detecting, Understanding, and Mitigating Unwanted Algorithmic Biascited by 319
  6. 2019Algorithms for Verifying Deep Neural Networkscited by 221
  7. 2017Fairness Testing: Testing Software for Discriminationcited by 126
  8. 2018Verisig: verifying safety properties of hybrid systems with neural network controllerscited by 84
  9. 2020Verification of Deep Convolutional Neural Networks Using ImageStarscited by 75
  10. 2021The Second International Verification of Neural Networks Competition (VNN-COMP 2021): Summary and Resultscited by 71
  11. 2020Learning Certified Individually Fair Representationscited by 69
  12. 2021Introduction to Neural Network Verificationcited by 63
6 more
  1. 2018Probabilistic Verification of Fairness Properties via Concentrationcited by 45
  2. 2019Perfectly Parallel Fairness Certification of Neural Networkscited by 42
  3. 2020Reachable Set Estimation for Neural Network Control Systems: A Simulation-Guided Approachcited by 36
  4. 2020Reachable Polyhedral Marching (RPM): A Safety Verification Algorithm for Robotic Systems with Deep Neural Network Componentscited by 36
  5. 2018Verification for Machine Learning, Autonomy, and Neural Networks Surveycited by 24
  6. 2021Removing biased data to improve fairness and accuracycited by 17

Built on by

Cited by year

Citing papers per year: 2024: 3; 2025: 3.

6 papers.

  1. 2024Solving Probabilistic Verification Problems of Neural Networks using Branch and Boundcited by 4
  2. 2025Justified Evidence Collection for Argument-based AI Fairness Assurancecited by 2
  3. 2024Knowledge Authoring with Factual English, Rules, and Actionscited by 1
  4. 2024Analyzing the Evolution of Graphs and Textscited by 0
  5. 2025Language Models That Walk the Talk: A Framework for Formal Fairness Certificatescited by 0
  6. 2025The Fair Game: Auditing & Debiasing AI Algorithms Over Timecited by 0